Energy Analysis of the New Dual-Pressure Low-Temperature

Jun 23, 2016 - In the case of C2+ fraction removal downstream of the dual-pressure low-temperature distillation process, a two-column layout has been ...
0 downloads 12 Views 2MB Size
Subscriber access provided by - Access paid by the | UCSB Libraries

Article

Energy Analysis of the New Dual Pressure Low-Temperature Distillation Process for Natural Gas Purification Integrated with NGLs Recovery. Stefano Langè, and Laura A. Pellegrini Ind. Eng. Chem. Res., Just Accepted Manuscript • DOI: 10.1021/acs.iecr.6b00626 • Publication Date (Web): 23 Jun 2016 Downloaded from http://pubs.acs.org on June 30, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Industrial & Engineering Chemistry Research is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Energy Analysis of the New Dual Pressure LowTemperature Distillation Process for Natural Gas Purification Integrated with NGLs Recovery. Stefano Langéa*, Laura A. Pellegrinia a

Politecnico di Milano, Department of Chemistry, Materials and Chemical Engineering “G.

Natta”, piazza L. da Vinci 32, I-20133, Milano, Italy. The global natural gas demand is expected to rise continuously over the next decades, compelling industries to study and develop new natural gas purification processes, such as lowtemperature ones, in order to allow the profitable exploitation of low-quality gas reserves (with high contents of CO2 and/or H2S) that up to some years ago were not considered suitable for their commercialization. When dealing with natural gas, not only acidic gases are present in the main stream, but also hydrocarbons heavier than methane (ethane, propane, n-butane, etc…). These substances may have to be removed from the produced gas for the dew point control of the sales gas or to manage its BTU content. Moreover, they are widely used for petrochemical productions and their recovery can be profitable for natural gas producers. In this work, an energy analysis is carried out in order to assess the energy requirements for the recovery of Natural Gas Liquids (NGLs) during natural gas purification at low-temperatures. The results are compared with the ones obtained when traditional processes (amine units +

ACS Paragon Plus Environment

1

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 100

dehydration + demethanizer) are used for the same purpose, in order to define the trade-off between the two technologies, extended to the overall production chain, under different possible feed stream compositions. The complete natural gas production chain has been studied. The energy requirements for the considered process units for both the traditional and the low – temperature schemes have been calculated by means of detailed process simulations in Aspen Hysys® V7.3 or by means of rules of thumb (when possible). Results prove the feasibility of low-temperature purification processes when dealing with gases having high amounts of acidic compounds and different concentrations of NGLs. 1. Introduction Forecasts on the global energy demand show a rapid increase worldwide over the next twenty years. Natural gas is the fossil fuel which shows the highest growth trend1 largely due to lower carbon dioxide emissions. Reports and studies made in the open literature on the distribution of currently known gas reserves establish that about 40% of the remaining ones are sour2 and about 30% present high CO2 contents3, typically between 15% and 80%4. H2S concentrations may reach the 15%5. These kinds of low-quality gas reserves are located in different geographic areas, such as South-East Asia, North Africa, Middle East, USA and Australia3. Examples of these gas fields with high concentrations of CO2 and/or H2S contents are shown in Table 1.

Table 1. Acidic gases contents in some natural gas reserves. Gas Field

Location

Acidic gases

Reference

Natuna

Indonesia

>70% CO2

3

Kapuni

New Zeland

43.8% CO2

5

Uch

Pakistan

46.2% CO2

5

ACS Paragon Plus Environment

2

Page 3 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

LaBarge

USA

65% CO2

6

Harweel Cluster

Oman

20% CO2

7

Lacq

France

15% H2S

5

Bearberry

Canada

90% H2S

7

Bujang

Malaysia

66% CO2

8

Sepat, Noring, Inas

Malaysia

60% CO2

8

Tangga Barat

Malaysia

32% of CO2

8

Ular, Gajah

Malaysia

50% of CO2

8

Beranang

Malasyia

28% of CO2

8

Bergading

Malaysia

40% of CO2

8

Palas NAG

Malaysia

46% of CO2

8

K5

Malaysia

70% of CO2

8

J5

Malaysia

87% of CO2

8

J1

Malasyia

59% of CO2

8

T3

Malaysia

62% of CO2

8

Tenggiri Mrn.

Malaysia

47% of CO2

8

Shah

United Emirates

Arab 23 % H2S, 10% 9 CO2

El Tapial

Argentina

72.24% of CO2

10

Molve

Croatia

23% of CO2

11

Kalinovac

Croatia

12.5 % of CO2

11

Traditional natural gas sweetening processes (chemical or physical absorption), in particular chemical scrubbing by means of aqueous alkanolamine solutions, can be too energy intensive when the acidic gases content in the feed stream is high12-14. In this scenario, attention has been devoted in the last decades to study and develop new natural gas purification processes,

ACS Paragon Plus Environment

3

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 100

particularly low-temperature ones, in order to decrease the overall production costs13, maintaining the required purity specifications on the produced gas, and to allow the profitable exploitation of sub-quality gas reserves that up to some years ago were not considered suitable for the market due to the higher costs related to the removal of high amounts of contaminants with traditional process solutions. Low-temperature processes for natural gas purification may operate at temperatures and pressures where a CO2 rich solid phase can form inside process equipment. Several solutions have been proposed in literature to handle carbon dioxide freezing. These processes can be designed to allow the formation of a solid phase in a dedicated zone or to completely avoid CO2 freezing. Low-temperature purification processes that allow the formation of a CO2-rich solid phase are: the CFZTM process15-20, the Cryocell® process21-22, the Cryopur® process23-24, the integrated cryogenic CO2 removal coupled with pressurized natural gas liquefaction25. Processes which avoid the freezing of carbon dioxide are: the Ryan-Holmes process26-28 where an extraction solvent (normally a hydrocarbon heavier than methane, such as n-butane) is used to shift CO2 freezing conditions at lower temperatures and pressures; the Sprex® process29-30 where a bulk removal of CO2 is performed in a pressurized low-temperature distillation column, operated away from freezing conditions, while the final purification of the gas is made with a traditional MDEA unit and, recently, a new patented dual pressure lowtemperature distillation process31,

14

designed to bypass the freezing point of CO2 in mixtures

with methane by means of a specific thermodynamic cycle. In this work, the dual pressure low-temperature distillation process31 has been taken into account, since it allows to perform the purification by means of a distillation unit made by common equipment of the process industry. The process allows to reach commercial purity

ACS Paragon Plus Environment

4

Page 5 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

avoiding solid phase formation, without the need of ad-hoc designed units to handle dry ice formation or the addition of a third component, the extraction solvent, that must be regenerated. Moreover, from low-temperature distillation processes, CO2 can be obtained in liquid phase under pressure, suitable conditions for EOR or CCS purposes. When performing the purification by means of low-temperature process distillation, the compounds with lower volatility than CO2 are collected in the bottom product, hence suitable purification technologies are needed to recover C2+. Conversely, when natural gas sweetening is performed by means of classical chemical scrubbing, ethane and heavier hydrocarbons are recovered in the sweet gas, since they are less soluble in the aqueous phase. NGLs have their own market and they can be used for energy production or as raw materials for petrochemical productions32; therefore their recovery can be of interest, depending on the overall process economics. The increasing production of NGLs is mostly related to the exploitation of associated natural gas32-33 and to the reduction of gas flaring practices. The content of C2+ in gas streams may vary greatly depending on the kind of the considered gas reserve; examples are listed in Table 2.

Table 2. C2+ contents in some gas reserves. Gas Field

Location

C2+ [%]

Kind of gas

Reference

Parentis

France

26.4

Associated

5

Ekofisk

Norway

14.4

Associated

5

Maracaibo

Venezuela

16.3

Associated

5

Uthmaniyah

Saudi Arabia

33.9

Associated

5

Burgan

Kwait

22.7

Associated

5

ACS Paragon Plus Environment

5

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 100

Kirkuk

Iraq

32.5

Associated

5

Ardjuna

Indonesia

28.9

Associated

5

El Tapial

Argentina

12.78

Associated

10

Bakken Lean

USA

27

Shale

34

Bakken Rich

USA

36.2

Shale

35

Rockies Lean

USA

23.5

Shale

34

Rockies Rich

USA

39

Shale

34

Permian/Delaware Lean USA

25.1

Shale

34

Permian/Delaware Rich USA

35.2

Shale

34

Haynesville Lean

USA

8.6

Shale

34

Haynesville Rich

USA

18.6

Shale

34

Molve

Croatia

6.7

Conventional

11

Kalinovac

Croatia

12.2

Conventional

11

When dealing with classical natural gas production processes, specific technologies, well known in the open literature35, are available. These technologies operate mostly the separation between C2+ and methane, without incurring in non-ideal VLE behaviors that occur when also acidic gases (CO2 and H2S) are present in the mixture. For low-temperature natural gas purification processes, non-ideal VLE behaviors, regarding basically CO2-C2H6 and hydrocarbons-H2S systems35, should be taken into account during process design for the efficient recovery of C2+. On the GPSA Handbook35 some possible process schemes have been proposed. Finn and O’Brien33 considered possible pathways for the purification of CO2-rich gas streams and the recovery of NGLs; in particular they discussed the feasibility of the application of the CFZTM technology to a demethanizer unit inside the four-column Ryan-Holmes process35 and considered the recovery of heavy hydrocarbons before the separation between methane and

ACS Paragon Plus Environment

6

Page 7 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

carbon dioxide. To break the azeotrope that occurs in the CO2-C2H6 system, an extraction solvent should be used to enhance the relative volatility between the two components35. Some authors3639

considered the performances of the extractive distillation to separate ethane and carbon

dioxide, studying the controllability, different configurations for energy savings and the effect of the selected extraction solvent on the separation process. In all these works, only a two-column configuration has been investigated. In the present work, several possibilities have been taken into account to perform the NGLs recovery. In particular two different configurations have been adopted and compared to break the CO2-C2H6 azeotrope: the standard two-columns extractive distillation has been compared with a three-columns scheme, where a preliminary distillation unit performs the bulk removal of CO2 or ethane from the main stream, depending on the global composition of the mixture respect to the azeotrope composition. Then, the analysis has been devoted to define whether it is energetically profitable to recover NGLs before or after the methane purification unit, depending on the inlet content of C2+ in the gas stream. Hence, energy requirements of the low-temperature distillation plant for natural gas production have been estimated adopting the net equivalent methane concept, as described by Pellegrini et al.40. Results have been compared with the overall energy requirements of a traditional natural gas production plant, where purification is performed by means of chemical scrubbing with MDEA. The trade-off between the two solutions has been defined, depending on the acidic gases content and on the NGLs content of the feed stream. The comparison has been carried out considering also the case when CO2 has to be compressed for EOR or CCS in order to better estimate the cost of CO2 recompression on the trade-off between the two processes.

ACS Paragon Plus Environment

7

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 100

Process configurations have been studied by means of rules of thumb, when possible, and by means of process simulations with Aspen Hysys® V7.341. The energy expenses in terms of net equivalent methane useful to supply energy to the overall production plant have been converted in terms of percentage of the produced gas that needs to be burned to drive the entire process forward. The study has been performed considering several possible compositions of the feed gas in terms of acidic gases (CO2 and H2S) and C2+, trying to cover the widest range of natural gas compositions. This work aims to provide a quantitative investigation on the relative trade – off, in terms of energy demand, between the two natural gas purification chains, considering different possible scenarios and different options for the low – temperature process layout. The investigation is performed by means of a model – based approach, considering the complete natural gas production chain (together with CO2 recompression) and all the most important process units involved. As for the authors knowledge, no quantitative and detailed comparisons between traditional purification chains and new low – temperature purification ones (considering all the required process units) is available in the literature and, typically, the works proposed on this topic are scarce or limited only to general considerations about the low – temperature purification scheme, without any direct quantitative analysis.

2. Thermodynamic considerations To perform an accurate process design and simulation, it is of paramount importance to set-up a proper and reliable thermodynamic framework that accurately describes phase equilibria. In low-temperature natural gas purification by means of distillation, not only VLE should be taken into account, since, at temperatures below 216.59 K42, the triple point of CO2, carbon dioxide can

ACS Paragon Plus Environment

8

Page 9 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

form a solid phase; hence SLVE must be calculated in order to define suitable operating conditions to avoid solid formation inside process equipment. In this work the SRK EoS43 and the CO2 Freeze out utility available in Aspen Hysys® V7.3 have been considered. The reliability of the selected thermodynamic framework has been tested by comparison with literatureavailable experimental data and results obtained by means of an in-house Fortran routine based on a classical approach for solid-fluid phase equilibria already used and tested in previous works44, 14. This routine calculates fluid phase fugacities with the SRK43 and PR45 EoSs and the solid phase fugacity from the Gibbs free energy of CO2 melting. The reliability of the thermodynamic framework has been already discussed in a previous work14 for the VLE and SLVE of CH4-CO2, CO2-H2S and CH4-H2S systems. In this work, the validation is presented for different binary systems of interest when dealing with multicomponent mixtures of natural gas. Prediction of solubility of solid CO2 in mixtures with hydrocarbon is discussed; CO2-C2H6, CO2C3H8 and CO2-n-C4H10 systems are taken into account (Figure 1). The effect of the nature of the hydrocarbon on the solubility of CO2 in the liquid phase is shown in Figures 1d – 1e, considering experimental data46 for binary systems of CO2 and light hydrocarbons from methane to n-butane and showing the effect of the addition of n-butane to a mixture of methane and carbon dioxide. Regarding VLE, the validation is made considering binary mixtures of CO2 and H2S with ethane, propane and n-butane (Figures 2-3). Experimental data for the solubility of solid CO2 in mixtures with hydrocarbons are taken from the work by Kurata46, while binary VLE data are taken from the NIST TDE database47. a)

ACS Paragon Plus Environment

9

Industrial & Engineering Chemistry Research

1

CO2-C2H6

0.9 0.8 0.7

Exp. Data kurata, 1974 Hysys Freeze-Out SRK PR

xCO2

0.6 0.5 0.4 0.3 0.2 0.1 0 100

120

140

160 T [K]

180

200

220

b) 1

CO2-C3H8

0.9 0.8 0.7 0.6 xCO2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 100

0.5

Exp. Data kurata, 1974 Hysys Freeze-Out SRK PR

0.4 0.3 0.2 0.1 0 110

130

150

170 T [K]

190

210

c)

ACS Paragon Plus Environment

10

Page 11 of 100

1

CO2-n-C4H10

0.9 0.8 0.7

Exp. Data kurata, 1974 Hysys Freeze-Out SRK PR

xCO2

0.6 0.5 0.4 0.3 0.2 0.1 0 130

150

170 T [K]

190

210

d) 1

xCO2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

0.9

Kurata, 1974 - CO2-CH4

0.8

Kurata, 1974 - CO2-C2H6

0.7

Kurata, 1974 - CO2-C3H8

0.6

Kurata, 1974 - CO2-n-C4H10

0.5 0.4 0.3 0.2 0.1 0 90.0

110.0

130.0

150.0 170.0 T [K]

190.0

210.0

e)

ACS Paragon Plus Environment

11

Industrial & Engineering Chemistry Research

1 0.9

Kurata, 1974 - CO2-CH4

0.8 Kurata, 1974 -CO2-CH4-n-C4H10

0.7 0.6 xCO2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 100

0.5 0.4 0.3 0.2 0.1 0 90.0

110.0

130.0

150.0 170.0 T [K]

190.0

210.0

Figure 1. Comparison among results obtained with the Hysys CO2 Freeze-Out utility, results obtained using a classical approach44 and experimental data46 for the Tx solubility diagrams of: a) CO2-C2H6, b) CO2-C3H8 and c) CO2-n-C4H10 systems. d) Effect of the nature of the hydrocarbon on the solubility of CO2 in the liquid phase and e) effect of the addition of n-butane on the solubility of CO2 in the liquid phase in a ternary system containing also methane.

The results obtained using the classic approach with the two cubic EoSs match very well and are in good agreement with the experimental data by Kurata46 at high and low concentrations (close to 0) of CO2 in the liquid phase, while larger discrepancies are present for molar fractions of CO2 in the liquid phase below 0.4. The results obtained with the CO2 Freeze-out utility available in Aspen Hysys® are in good agreement with the ones obtained with the Fortran routine for the CO2-C2H6 system (Fig. 1a), while the difference increases for the CO2-C3H8 and CO2-nC4H10 systems. The results obtained with the CO2 Freeze-out utility are closer to the

ACS Paragon Plus Environment

12

Page 13 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

experimental points for values of the CO2 molar fractions in the liquid phase between about 0.10.5, where the classic approach shows higher deviations. For each value of the CO2 concentration in the liquid phase, the CO2 Freeze-Out utility overestimates the freezing temperature, remaining more conservative. Considering the nature of the hydrocarbon (Figure 1d), the shape of the solubility curve on the Tx diagram, for binary mixtures, is not significantly affected, because the nature of the solvent is similar. However, between 190 K and 210 K, differences on xCO2 at which freezing occurs are more evident among the different binary systems. The presence of heavier hydrocarbons in the natural gas stream, when performing lowtemperature distillation for natural gas purification, enhances the solubility of dry ice in the liquid phase, shifting freezing points of CO2 (SLVE line on a PT diagram) at lower temperatures and pressures. This effect is more evident on the SLVE locus as highlighted also by Ryan and Holmes28, GPSA35 and by Pellegrini et al.48. This is due to the enhancement of the bubble point of the mixture, that shifts the VLE region away from the SLE conditions. Considering the ternary system CO2 – CH4 – n - C4H10 (Figure 1e), it is possible to notice that, at a given temperature, the xCO2 at which freezing occurs is higher and this effect is particularly evident in the region of temperatures at which the maximum pressure of the SLVE locus of the CH4 – CO2 system occurs (for more details about the PT diagram of this binary system refers to the previous work14). Hence, the presence of n-butane limits the dry ice formation.

a)

ACS Paragon Plus Environment

13

Industrial & Engineering Chemistry Research

4

223.1 K 243.1 K 263.1 K

3.5

Exp. 223.1 K Exp. 243.1 K Exp. 263.1 K

CO2-C2H6

3 P [MPa]

2.5 2 1.5 1 0.5 0 0

0.2

0.4

xCO2

0.6

0.8

1

b) 7

CO2-C3H8

6 5 P [Mpa]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 100

4

277.6 K

294.3 K

273 K

252 K

266 K

244 K

230 K

Exp. 277.6 K

Exp. 294.3 K

Exp. 273 K

Exp. 252 K

Exp. 266 K

Exp. 244 K

Exp. 230 K

3 2 1 0 0

0.2

0.4 xC H 0.6 3 8

0.8

1

c)

ACS Paragon Plus Environment

14

Page 15 of 100

6

CO2-n-C4H10

5 4 P [MPa]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

292.6 K

277.9 K

270 K

250 K

227.98 K

Exp. 292.6 K

Exp. 277.9 K

Exp. 270 K

Exp. 250 K

Exp. 227.98 K

3 2 1 0 0

0.2

0.4

0.6

0.8

1

xn-C4H10 Figure 2. Comparison between results obtained with the SRK EoS43 and experimental data taken from the NIST TDE database47 for the VLE of binary CO2-hydrocarbons systems at different temperatures. Pxy diagrams of: a) CO2-C2H648, b) CO2-C3H8 and c) CO2-n-C4H10 systems.

The representation of the VLE of binary CO2-hydrocarbons systems is satisfactory: model results are in good agreement with experimental data, also for the azeotropic behavior of the CO2-C2H6 system. The mixture presents a minimum azeotrope, the composition of which does not vary significantly with pressure. This is the reason why extractive distillation is needed to break the azeotrope and separate CO2 and ethane.

a)

ACS Paragon Plus Environment

15

Industrial & Engineering Chemistry Research

4 3.5

P [MPa]

3

283 K

255 K

227 K

199 K

Exp. 283 K

Exp. 255 K

Exp. 227 K

Exp. 199 K

C2H6-H2S

2.5 2 1.5 1 0.5 0 0

0.2

0.4

0.6

0.8

1

xC2H6 b) 400

2.758 MPa 0.689 MPa Exp. 2.758 MPa Exp. 0.689 MPa

350

2.068 MPa 0.345 MPa Exp. 2.068 MPa Exp. 0.345 MPa

1.379 MPa 0.138 MPa Exp. 1.379 MPa Exp. 0.138 MPa

C3H8-H2S T [K]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 100

300

250

200 0

0.2

0.4

0.6

0.8

1

xC3H8 c)

ACS Paragon Plus Environment

16

Page 17 of 100

9 366.45 K 380.35 K Exp. 376.45 K

8 7

376.45 Exp. 366.45 K Exp. 380.35 K

n-C4H10-H2S

6 P [Mpa]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

5 4 3 2 1 0 0

0.2

0.4 0.6 xn-C4H10

0.8

1

Figure 3. Comparison between results obtained with the SRK EoS43 and experimental data taken from the NIST TDE database47 for the VLE of binary H2S-hydrocarbons systems: a) Txy diagram for H2S-C2H648, b) Pxy diagram for H2S -C3H8 (Pellegrini et al., 2015b) and c) Pxy diagram for H2S -n-C4H10.

The VLE of binary H2S-hydrocarbons mixtures is well reproduced by the thermodynamic model, except for the H2S-n-C4H10 system close to the critical conditions, where deviations between calculation results and experimental data are significant. Hydrogen sulfide and ethane does not form an azeotrope, but, at high concentrations of ethane, the relative volatility is greatly reduced. The H2S-C3H8 system presents a minimum azeotrope at low molar fractions of propane. Also for this system, the composition of the azeotrope does not change significantly with pressure. So, the separation of hydrogen sulfide from the main hydrocarbon stream via distillation is difficult. Finn and O’Brien33, according to the GPSA handbook35, suggested a

ACS Paragon Plus Environment

17

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 100

preliminary removal of H2S and C2+ from the main natural gas stream by means of extractive distillation. The thermodynamic framework has proven to be reliable and has been used for process simulations.

3. Description of process layouts and calculation methods The global natural gas purification process has been considered. Each unit of the main process has been studied in order to determine the overall energy requirements. In this work the following main assumptions have been considered: •

H2S is removed before the low-temperature processes for natural gas purification;



the final product is pipeline-quality gas: CO2 < 2mol%; H2S < 4 ppm49-50;



maximum hydrocarbons recovery is required;



C2+ are considered as ethane during process simulations;



produced ethane has a molar purity > 99%;



CO2 quality is for pipeline transmission51.

H2S removal before the low-temperature purification has been chosen because of its toxicity, according to the suggestion of Turton et al.52, its VLE behavior in mixtures with hydrocarbons and its SLVE behavior in mixtures with CO2 and CH4, as discussed by Pellegrini et al.48 and Langé et al.53. Heavier hydrocarbons have been lumped as ethane for the sake of simplicity. It has to be pointed out that, usually, the recovery of the C2+ fraction is not mandatory and can be considered on the basis of economic considerations and of transportability of the final product (dew point

ACS Paragon Plus Environment

18

Page 19 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

control or BTU content management). In this work, the C2+ fraction has been assumed to be always recoverable in order to extend the generality of the study. The inlet raw natural gas stream has been considered to be available at 50 bar, with a composition varying for H2S from 0 to 15 mol%, for CO2 from 5 to 65 mol% and for C2+ from 0 to 25 mol%, while the methane content results from the balance. The energy requirements of each process unit have been calculated in terms of equivalent methane40. The different quality of the energy has been considered, assuming to produce electric energy to drive compressors by means of a methane-fired combined cycle, to produce steam by means of a methane-fired boiler and to produce cooling duties by means of vapor-compression refrigeration cycles. Internal energy that can be recovered (such as heat at high temperature) has been considered as a methane saving. The overall energy performance of the plant has been calculated, in this way, as the percentage of the produced gas that has to be burned to supply energy to the overall purification plant. Useful parameters to perform equivalent methane calculations are reported in Table 3, while the relevant procedure is described below.

Table 3. Parameters for the calculation of the equivalent methane40. Parameter

Symbol

Methane Lower Heating Value

Value



50 MJ kg-1

Boiler Efficiency



0.80

Combined Cycle Efficiency



0.55

Second Law Efficiency



0.60

Ambient Temperature



298.15 K

ACS Paragon Plus Environment

19

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60



Page 20 of 100

Useful heat , needed to heat a process stream above the ambient temperature, is supplied by means of LP steam produced by a methane-fired boiler, that burns a methane mass flow equal to   with an efficiency equal to  :   = 





(1)

 

Cooling duty  , needed to cool a process stream below the ambient temperature, is produced by means of a vapor compression refrigeration cycle, having a coefficient of performance  , that burns a methane mass flow equal to   to drive cycle compressors that require a mechanical power   . The mechanical power can be calculated according to Eq. (2), where the actual value of the  can be calculated starting from the ideal one of a Carnot cycle (Eq.(3)) and the Second Law efficiency  (Eq.(4)).  !"# $ %&

 =

,() =

(2)

*

(3)

, + - .*/ ,

01

 = 0 •

(4)

1,23

The mechanical power is produced by means of a methane-fired combined cycle, having an efficiency equal to  :  = 4

$ %&

(5)

 



The equivalent methane for the production of cooling duties is:   = 0

 !"#

(6)

1  

ACS Paragon Plus Environment

20

Page 21 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research



Mechanical power   , needed to drive process machineries, is supplied by means of a methane-fired combined cycle with an efficiency equal to  , that burns a methane mass flow equal to   :   =

$ %&

(7)

 

Methane saving terms are represented by heat flows that can be recovered inside the process and mechanical powers produced by process streams expansion in turbines. For each block of the Block Flow Diagrams (BFDs) shown in Figs. 4-5, the net equivalent methane has been calculated in terms of kg s-1 per kmol h-1 of feed stream to the process unit. In the following sub-sections, each process unit is described and the energy terms accounted for the analysis are discussed.

3.1 Global BFDs of the considered natural gas production plants In this work, two complete natural gas production plants have been considered: a classic scheme (Fig. 4), composed of an acid gas removal unit, a dehydration unit and a demethanizer, and a low-temperature scheme (Fig. 5) that can be organized according to two different ways. In both these ways, H2S is selectively removed at the beginning and the desulfurized stream is then dehydrated before entering the cold part of the process. In the first pathway (Fig. 5a), the C2+ fraction is removed before the low-temperature dual pressure distillation process for the CH4/CO2 split, while in the second one (Fig. 5b) the C2+ fraction is separated after the lowtemperature dual pressure distillation unit. In this work the different processes have been compared considering also the CO2 recompression unit, that differs between the classic scheme (Fig. 4b) and the low-temperature

ACS Paragon Plus Environment

21

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 100

one (Figs. 5c–5d) according to the operating conditions at which the CO2 stream is obtained. This part will be discussed more in detail in the next sections. The BFDs of the two considered natural gas production plants are presented in Figs. 4-5.

a) Raw Gas Feed CH4/CO2/C2+/H2S/H2O

First MDEA sweetening unit

CH4/C2+/ H2O/ CO2

Second MDEA sweetening unit

CH4/C2+/ H2O

Demethanizer CH4/C2+

H2O

CO2

H2S

TEG dehydration unit

Sales Gas

C2+

b) Raw Gas Feed CH4/CO2/C2+/H2S/H2O

First MDEA sweetening unit

CH4/C2+/ H2O/ CO2

Second MDEA sweetening unit

CH4/C2+/ H2O

TEG dehydration unit

Demethanizer CH4/C2+

Sales Gas

Wet CO2 H2S

H2O

CO2 Compression and Dehydration

H2O

C2+

CO2

Figure 4. Classic scheme for the overall natural gas purification chain: a) without CO2 recompression and b) with CO2 recompression.

a) Raw Gas Feed CH4/CO2/C2+/H2S/H2O

MDEA selective H2S removal

H2S

CH4/CO2/ C2+/H2O

TEG dehydration unit

H2O

CH4/CO2 /C2+

C2+ Split

C2+

CH4/CO2

Dual Pressure Low-Temperature Distillation Unit

Sales Gas

CO2

b)

ACS Paragon Plus Environment

22

Page 23 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Raw Gas Feed CH4/CO2/C2+/H2S/H2O

MDEA selective H2S removal

CH4/CO2/ C2+/H2O

TEG dehydration unit

CH4/CO2 /C2+

Dual Pressure Low-Temperature Distillation Unit

H2O

H2S

C2+ Split

C2+/CO2

Sales Gas

CO2

c) Raw Gas Feed CH4/CO2/C2+/H2S/H2O

MDEA selective H2S removal

CH4/CO2/ C2+/H2O

TEG dehydration unit

CH4/CO2 /C2+

C2+ Split

CH4/CO2

Dual Pressure Low-Temperature Distillation Unit

Sales Gas

CO2 H2O

H2S

C2+ CO2 Pump

CO2

d) Raw Gas Feed CH4/CO2/C2+/H2S/H2O

MDEA selective H2S removal

CH4/CO2/ C2+/H2O

TEG dehydration unit

CH4/CO2 /C2+

Dual Pressure Low-Temperature Distillation Unit

C2+/CO2

C2+ Split

C2+

CO2 H2S

H2O

Sales Gas CO2 Pump

CO2

Figure 5. Low-temperature scheme for the overall natural gas purification chain: a) with C2+ recovery before the CH4 / CO2 separation by means of the dual pressure low-temperature process and without CO2 recompression; b) with C2+ recovery after the CH4 / CO2 separation by means of the dual pressure low-temperature process and without CO2 recompression; c) with C2+ recovery before the CH4 / CO2 separation by means of the dual pressure low-temperature process and with CO2 recompression and d) with C2+ recovery after the CH4 / CO2 separation by means of the dual pressure low-temperature process and with CO2 recompression.

ACS Paragon Plus Environment

23

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 100

The overall energy consumptions of each block of the BFDs and the relevant specific net equivalent methane (kg s-1 per kmol h-1 of feed stream) can be calculated starting from material balances. The net specific equivalent methane per each block has been parametrized as mathematic functions of simulations results when rules of thumb cannot be used. Rules of thumb are available for the MDEA unit14 and for the TEG dehydration unit54,

49

. The energy

performances of each process scheme have been evaluated as percentage of the produced gas to be burned in order to provide energy to the overall purification chain, according to: %6789:; =

DEFGH

+∑2IJ

2 @AA2 ,BC /

KL!#

× 100

(8)

Furthermore, the detail of the process units corresponding to the block in Figs. 4-5, the modeling and the methodologies adopted to calculate their relative energy expenses are discussed.

3.2 MDEA sweetening unit Chemical absorption by means of aqueous solutions of MDEA is the classic scheme (Fig. 4) considered for the removal of acidic gases (CO2 and H2S) from the main gas stream. Due to the selectivity of the solvent, this unit has been adopted also for the removal of H2S in the lowtemperature process (Fig. 5). This choice has been already discussed (see Section 3). Moreover, since H2S is fed to Sulfur Recovery Units (SRUs) at about 1 – 2 barg, the choice of a preliminary removal by means of selective chemical absorption allows to obtain H2S for the SRU unit at the desired pressure, without complicated separations from the main stream (strong non-idealities in the VLE behavior of mixtures of H2S with hydrocarbons and CO2). The main energy consumption of the absorption unit is the heat required by the reboiler of the regeneration column55. The amount of heat necessary for the regeneration of the solvent is

ACS Paragon Plus Environment

24

Page 25 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

mainly due to the high heat of vaporization of water (that is the compound of the solvent mixture present in largest amounts, especially on molar basis) and to the heat required to break chemical bonds formed during acidic gases absorption. A Process Flow Diagram (PFD) for the MDEA unit is reported in Fig. 6. The calculation of energy consumptions related to this process unit has been performed according to a rule of thumb14 which considers an overall LP steam consumption at the reboiler of the regeneration column equal to 0.14 kg per L of lean circulating solvent. The validity of this assumption has been discussed in a previous work14. Once calculated the volumetric flowrate  of the circulating lean solvent, the amount of heat required by the regeneration column is: T = 0.14 RS

(9)

T where RS is the heat of vaporization of water at about 3.5 bar14.

ACS Paragon Plus Environment

25

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 100

Figure 6. PFD of a classic MDEA unit. → Heat required by the reboiler, accounted as key parameter for the net equivalent methane calculations (Eq. (1)).

The unit consists of an absorption column, where the gas feed is sweetened by contacting in counter-current the lean solvent, and of a regeneration column, where acidic gases are rejected at the top of the column and the lean regenerated solvent is recycled to the process. A proper makeup of the solvent is needed. In order to save energy, the sensible heat of the hot lean solvent, coming from the reboiler of the regeneration column, is recovered by means of a cross heat exchanger, heating the rich amine stream (loaded with the acidic compounds) to be fed to the regeneration column. The absorption is operated under pressure, while, in order to favor the recovery of acidic gases, the regeneration is operated at low-pressures, typically about 1-2 barg. For the removal of CO2 and H2S (Fig. 4) two units in series are used, while, when only H2S removal is needed (Fig. 5), only one unit is required. The sweet gas from the top of the absorber is saturated with water. The water content in this gas stream is affected by the presence of CO2 and H2S, which allow a larger amount of water at saturation conditions as discussed in the GPSA Handbook35. The amount of water in the produced sweet gas has been calculated by means of process simulations with Aspen Hysys® V7.3, using the DBR Amine Package. Calculations have been performed considering, for the absorption column, the same configuration reported in the work by Langé et al.14. For the classic scheme (Fig. 4), the inlet content of CO2 has been varied from 5 to 65 mol%, while the one of H2S from 0 to 15 mol%. The absorber pressure has been set at 50 bar, the same of the natural gas feed. For this process layout, acidic gases should be removed as indicated in Section 3, thus, since their content in the sweetened gas is much lower than the one of hydrocarbons, the water

ACS Paragon Plus Environment

26

Page 27 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

content is not affected by their presence and it results to be constant and equal to 0.0018 mol/mol, according to the operating conditions of the top tray of the absorber (50 bar and 40 – 41 °C). When considering the selective removal of H2S for the second process scheme (Fig. 5), the presence of CO2 in the remaining gas is not negligible and affects the content of water in the produced gas35 from the top of the absorber. For this reason, process simulations have been performed in order to define the variation of the molar fraction of water with the inlet content of acidic gases (CO2 and H2S). Conversely, the number of absorber trays has been varied in order to meet the specification on the final H2S content. Since H2S is removed to the level of ppms, once the temperature and the pressure are fixed, the molar fraction of water in the produced gas is affected only by the presence of CO2. The obtained results from absorber simulations have been correlated. The final water content in the desulfurized gas has been expressed as a function of the CO2 mol% of the feed gas (Eq. (10)). VWX UT = 6.267 × 10.\ %T@AA + 1.759 × 10.`

(10)

The graphic representation of the reliability of Eq. (10) is shown in Fig. 7.

ACS Paragon Plus Environment

27

Industrial & Engineering Chemistry Research

0.0023 0.0022 yH2O Outlet Gas

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 100

0.0021 0.002 0.0019 0.0018 0.0017 0.0016 0.0015 5

10 15 20 25 30 35 40 45 50 55 60 65 Feed CO2 mol%

Figure 7. Correlation (Eq. (10)) of the absorber simulation results (Aspen Hysys V.7.3, DBR Amine Package) for the desulfurization of the inlet feed gas by means of chemical absorption of H2S in a 40 wt% MDEA aqueous solution at 50 bar (gas outlet temperature equal to 41 °C).

The obtained correlation allows a good representation of the water content of the desulfurized gas. It has to be pointed out that, for the second process layout (Fig. 5), where no H2S is present in the gas feed, Eq. (10) has been still adopted to calculate the amount of water in the raw gas feed. This is correct because, after desulfurization, the gas is saturated with water according to the operating conditions of the top absorber tray.

3.3 TEG dehydration unit For the removal of water before the demethanizer unit (Fig. 4) and the low-temperature purification (Fig. 5), Triethylene Glycol (TEG) has been considered as solvent for gas

ACS Paragon Plus Environment

28

Page 29 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

dehydration. The configuration of the TEG dehydration process is similar to the one of the MDEA sweetening unit (Fig. 8).

Figure 8. PFD of a classic TEG Dehydration unit. → Heat required by the reboiler, accounted as key parameter for the net equivalent methane calculations (Eq. (1)).

The wet natural gas is contacted counter-currently with the lean regenerated TEG in an absorption column. The produced gas is dry, while the solvent, loaded with water, is sent to a regeneration unit. The regenerator separates water at the top and TEG at the bottom. Hot TEG is

ACS Paragon Plus Environment

29

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 100

then pumped (lower viscosity), cooled in a cross heat exchanger, furthermore cooled by heat exchange with the produced dry gas and recycled to the top of the absorption column. Absorption occurs under pressure, while regeneration is operated at low pressure, close to the atmospheric one. The limit for the temperature at the reboiler of the regeneration section is given by the maximum allowable temperature at which the solvent is stable. TEG is chosen because of its high decomposition temperature54,

49

, its high boiling point and because it is easier to be

efficiently regenerated at atmospheric pressure than other glycols49. Heat required for TEG regeneration is the most important term of energy requirement in the dehydration unit. The calculations of the heat duty required by the reboiler of the TEG regeneration column have been made adopting rules of thumb, as suggested by Maddox54 and Mokhatab et al.49. Maddox54 suggested to adopt as energy consumption at the reboiler a duty equal to 560 kJ per L of circulating TEG. Mokhatab et al.49 suggested to calculate the circulating TEG in the range 2–6 US gal of TEG per lb of removed water and to increase this value with a safety margin of 10– 30%. In this work, 3 US gal of TEG per lb of water and a safety factor of 30% have been considered to calculate the circulating solvent.

3.4 Demethanizer The demethanizer unit is used in the classic purification scheme (Fig. 4) to remove the C2+ fraction from a sweetened and dehydrated natural gas stream. The unit performs the separation between methane and heavier hydrocarbons in order to avoid possible retrograde condensation of the mixture during transportation. The PFD of this process unit is shown in Fig. 9. Several possible configurations exist for this process. In this work the PFD has been defined starting from literature-available PFDs56.

ACS Paragon Plus Environment

30

Page 31 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Figure 9. PFD of the demethanizer unit. → cooling duty required by low-temperature heat exchangers; → mechanical power required by compressors and → heat required by the reboiler. Only cooling duties and mechanical power are accounted as key parameter for equivalent methane calculations (Eqs. (6-7)).

The unit consists of two sections: a cold box and a distillation column. In the cold box, cold recovery between the produced gas and the feed stream is realized. An external cooling duty is required to furthermore cool the feed stream down to the desired temperature level, obtaining a mixed vapor–liquid stream. The vapor fraction of the cold feed is separated into two streams: one gas stream is furthermore cooled by the distillation column product stream, expanded and fed to the top of the distillation unit to provide totally or part of the liquid reflux. The other stream is

ACS Paragon Plus Environment

31

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 100

expanded in a turboexpander and fed on the fifth tray from the top of the column56. The liquid portion of the main feed stream after the first cooler is expanded and sent to the eighteenth tray from the top. The gas feed has been considered at 25 °C, 50 bar. The feed has been considered as a binary CH4-C2H6 mixture and its composition has been varied from 5 to 75 mol% of C2H6. The cold box operates at 50 bar. The distillation column has 30 theoretical trays and is operated at 25 bar, as suggested by Luyben56. A condenser is placed at the top of the distillation unit in order to guarantee a final methane content in the produced gas higher than 99 mol%, while the bottom product has 100 ppm of CH4, in order to enhance the methane recovery. The final produced gas is re-compressed to 50 bar. The minimum approach in the heat exchangers has been set to 5 °C. The process unit has been simulated with Aspen Hysys® V7.3, using the SRK EoS43. Since the reboiler does not require heat at high temperatures, its contribution to the net equivalent methane has been neglected, while cooling duties at low-temperatures and mechanical power for compressors have been accounted as key-parameters for the energy analysis. Per each inlet composition, the process operating conditions have been optimized in order to find the minimum energy required to perform the separation according to the purity specifications. The vapor fraction of the cooled feed stream after the cold box (that determines the precooling temperature of the feed stream) and the split factor of the vapor portion of the feed stream have been considered as degrees of freedom for the optimization. In particular, the optimal split factor varies from 0.1 (feed stream having 5 mol% of C2+) to 0.8 (feed stream having 75 mol% of C2+) and the optimal vapor fraction of the cooled feed stream after the cold box varies from 0.96 (feed stream having 5 mol% of C2+) to 0.11 (feed stream having 75 mol% of C2+). It has to be pointed out that, for high amounts of ethane (above 15 mol%) in the inlet feed, it is not possible to guarantee the final methane purity in the produced gas providing the reflux only by means of the

ACS Paragon Plus Environment

32

Page 33 of 100

expanded portion of the gas feed. The results of process simulations have been used to parametrize the specific net equivalent methane of this process unit as function of the inlet molar fraction of ethane: 4bc = 7.419 × 10.d eT\ + 3.104 × 10.d a,A

(11)

The reliability of the proposed correlation is shown in Fig. 10.

0.0001 CH4,EQ. [(kg/s)/(kmol/h)]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

0.00009 0.00008 0.00007 0.00006 0.00005 0.00004 0.00003 0.00002 0.05

0.15

0.25

0.35 0.45 z C2H6

0.55

0.65

0.75

Figure 10. Reliability of the correlation (Eq. (11)) to represent the specific net equivalent methane as function of the molar fraction of ethane in the feed stream to the demethanizer.

The agreement is good, except for the point at 15 mol% of ethane in the inlet feed stream. For feed compositions below this value, the reflux ratio of the distillation column provided by the external condenser is zero and the reflux is produced only by using the portion of the expanded feed stream, which enters the column at the first tray from the top. In these conditions, the optimization has been constrained (purity equal to 99 mol%) in order to find the set of the

ACS Paragon Plus Environment

33

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 100

degrees of freedom that allow to obtain the minimum energy consumptions together with the required purity of the final produced gas.

3.5 CO2 recompression and dehydration Carbon dioxide rejected from the MDEA unit is obtained at the top of the regeneration column at a temperature of about 30 °C14 and is typically available at about 1 – 2 barg and saturated with water. To transport CO2 for EOR or CCS, the stream should be dehydrated and compressed to a pressure of about 130 bar57. The water content of the produced CO2 has been estimated through process simulations for the MDEA unit performed with Aspen Hysys® V7.3, using the DBR Amine Package. At 2 bar and 30 °C this value is 0.0216 mol/mol. The recompression and dehydration process has been simulated using Aspen Hysys® V7.3 and the SRK EoS43. The PFD of the compression and dehydration train adopted for the classic scheme (Fig. 4) is shown in Fig. 11.

Figure 11. PFD of the CO2 compression and dehydration unit. → mechanical power required by compressors, accounted as key parameter for equivalent methane calculations (Eq. (7)).

ACS Paragon Plus Environment

34

Page 35 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

The inlet wet CO2 stream is fed at 30 °C and 2 bar to a four-stage intercooled compression train. The temperature at the outlet of intercoolers has been set at 30 °C and pressure drops have been neglected. The outlet pressure, Pn, from the n-th compression stage has been calculated according to: g = g.* + where

0Ghi 02j

J

0Ghi j / 02j

(12)

is the compression ratio between the inlet and outlet pressure of the fluid in the

total compression train. The condensed water (with traces of CO2) is separated after each intercooler and expanded to the atmospheric pressure. The specific work required by the unit is 3.95 kW per kmol h-1 of inlet wet CO2 feed stream. For the calculation of the net equivalent methane, only mechanical power has been considered for this unit (Eq. (7)).

3.6 Dual pressure low-temperature distillation process The dual pressure low-temperature distillation process proposed by Pellegrini31 has been considered for the separation of methane from CO2 and CO2+C2H6. In this work, only the optimized layout has been taken into account for the energy evaluation while the detailed description of the process can be found elsewhere31, 14. The PFD of the process unit is shown in Fig. 12.

ACS Paragon Plus Environment

35

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 100

Figure 12. PFD of the Dual Pressure Low-Temperature Distillation process31. → cooling duty required by low-temperature condensers; → heat flow at low temperature required by the intermediate heater. Only cooling duties are accounted as key parameters for equivalent methane calculations (Eq. (6)). The value of COPf for the refrigeration cycle is 0.6714.

The process involves two sections of a distillation unit, operated at 50 and 40 bar respectively. The produced gas is methane at high purity, while heavier compounds are collected in the bottom product. Simulations have been performed in Aspen Hysys® V7.3 with the SRK EoS43 in order to define mathematical functions that relate the specific net equivalent methane to the inlet gas feed composition. The inlet gas is available at 50 bar at its dew point. Binary mixtures of CH4CO2 and ternary mixtures of CH4-CO2-C2H6 have been considered in order to cover all the possible cases. When the C2+ fraction is removed upstream of this process unit, the specific net

ACS Paragon Plus Environment

36

Page 37 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

equivalent methane is function only of the CO2 content of the inlet gas stream. (Eq. (13)), while, when the C2+ separation is performed downstream of the methane removal, the function depends also on the C2H6 content of the feed stream (Eq. (14)). For the first case, simulations have been performed changing the CO2 molar fraction in the feed stream from 5 up to 90 mol%, while, for the second case, the C2H6 content has been varied from 0 up to 25 mol% and the CO2 content from 5 to 65 mol%. For the calculation of the net equivalent methane only cooling duties have been considered, since the amount of heat required by the intermediate heater is at lowtemperatures and the process feed stream can be used to provide it. Since the reboiler temperatures are close to the ambient one, no LP steam or other expensive heat sources are required. \ d 0k a = −6.955 × 10.` eT + 2.299 × 10.T eT − 3.0 × 10.T eT + 1.933 × a,A ` T 10.T eT − 6.251 × 10.` eT + 1.038 × 10.` eT + 4.1 × 10.d

(13)

0k a,A = \ d ` a T =neT\ oeT + pneT\ oeT + qneT\ oeT + ;neT\ oeT + :neT\ oeT +

rneT\ oeT + sneT\ o

(14)

The parameters a, b, c, d, e, f, g of Eq. (14) have been defined as functions of the inlet molar fraction of ethane according to: a T d ` t = u* eT\ + uT eT\ + u` eT\ + ua eT\ + ud eT\ + u\ t = =, p, q, ;, :, r, s (15)

The values for the parameters K1 – K6 are reported in Table 4 together with their accuracy R2.

Table 4. Parameters used for the correlation adopted in Eq. (15) and their accuracy R2. θ

K1

K2

K3

K4

K5

K6

R2

ACS Paragon Plus Environment

37

Industrial & Engineering Chemistry Research

a

2700.773

306.608

-20.926

0.562

6.701E-03

b

-5959.493 3330.620

-689.066

49.108

-1.592

-4.539E-03 1

c

5110.347

606.635

-45.760

1.804

-8.748E-03 1

d

-2141.973 1208.280

-263.909

21.402

-1.031

1.145E-02

e

449.733

-254.900

58.102

-5.117

0.303

-4.815E-03 1

f

-41.867

23.400

-5.563

0.513

-0.039

9.200E-04

1

g

0.427

-5.333E-02

-2.667E-03 7.333E-03

1.164E-03

4.400E-05

1

-1505.447

-2867.227

1

1

The reliability of the proposed correlations is shown in Fig. 13 for Eq. (13) and in Fig. 14 for Eq. (14).

2.50E-04

CH4, EQ. [(kg/s)/(kmol/h)]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 38 of 100

2.00E-04 1.50E-04 1.00E-04 5.00E-05

0.00E+00 0.05

0.15 0.25 0.35 0.45 0.55 z CO2

0.65 0.75 0.85

Figure 13. Reliability of the correlation (Eq. (13)) to represent the specific net equivalent methane as function of the molar fraction of carbon dioxide in the feed stream when ethane is removed upstream of the dual pressure low-temperature distillation process.

ACS Paragon Plus Environment

38

Page 39 of 100

6.00E-04 5.00E-04 CH4, EQ. [(kg/s)/(kmol/h)]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

25%C2+

20%C2+

15%C2+

10%C2+

5%C2+

0%C2+

4.00E-04 3.00E-04 2.00E-04 1.00E-04 0.00E+00 0.05

0.15

0.25

0.35 z CO2

0.45

0.55

0.65

Figure 14. Reliability of the correlation (Eq. (14)) to represent the specific net equivalent methane as function of the molar fraction of carbon dioxide and ethane in the feed stream when ethane is removed downstream of the dual pressure low-temperature distillation process.

The agreement of the proposed correlations with simulation results is quite satisfactory.

3.7 Two-column extractive distillation for the removal of NGLs before Dual Pressure LowTemperature Distillation for methane purification To perform the removal of the C2+ fraction from the natural gas stream, a first analysis regards a two-column extractive distillation unit to break the CO2-C2H6 azeotrope and extract NGLs before the CH4-CO2 separation. The PFD of the proposed solution is shown in Fig. 15.

ACS Paragon Plus Environment

39

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 40 of 100

Figure 15. PFD of the two-column extractive distillation unit for the separation of the C2+ fraction upstream of the Dual Pressure Low-Temperature Distillation process. → cooling duty required by low-temperature condensers; → heat flow required by reboilers; → mechanical power required by the compressor; → amount of useful heat at high temperature that can be recovered inside the process (net equivalent methane savings). For the specific net equivalent methane calculation, cooling duties have been calculated according to Eq. (6), heat flows have been calculated according to Eq. (1) and mechanical power has been calculated according to Eq. (7).

The inlet gas feed is available at 50 bar and at 25 °C. Due to the presence of a minimum azeotrope in the CO2-C2H6 system, an extraction solvent (n-C4H10) has been adopted in order to

ACS Paragon Plus Environment

40

Page 41 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

break the azeotrope and separate the C2+ fraction of the feed stream. Normal butane has a critical pressure of about 38 bar58, hence, to perform the separation, the operating pressure should be lower than such a value. So, the feed stream is expanded to 35 bar. The level of pressure is chosen in order to remain below 38 bar and, at the same time, to keep the pressure sufficiently high to avoid excessively low temperatures at column condensers and higher pumping expenses for CO2 recompression. The process has been studied with Aspen Hysys V7.3®, using the SRK EoS43. The two distillation columns have 30 theoretical trays. The extraction solvent is fed at 10 °C on the fourth tray from the top of the first distillation column. Methane and carbon dioxide are recovered in gas phase as top product stream from the first distillation column, while ethane and n-butane at the bottom in liquid phase. The produced gas stream is compressed back to 50 bar and is sent to the Dual Pressure Low-Temperature Distillation process, while the extraction solvent is regenerated in a second distillation column. The extraction solvent flow-rate is chosen as the minimum value that allows to minimize the content of carbon dioxide in the bottom stream of the first distillation column and to maximize the recovery of CO2 in the produced gas. Due to the high boiling point of n-butane under pressure, the temperatures at the reboilers of the two distillation units are higher than 100 °C, and so LP steam is needed. The sensible heat of the regenerated solvent can be recycled to the process. The inlet composition of the feed stream has been varied in order to cover the range of interest: 5–25 mol% of ethane and 0-65 mol% of carbon dioxide, while methane is the balance. Results of process simulations in terms of specific net equivalent methane have been correlated by means of the following expressions as function of the feed composition of ethane and carbon dioxide:

ACS Paragon Plus Environment

41

Industrial & Engineering Chemistry Research

Av a,A = wneT\ o:xyn6neT\ oeT o

(16)

` T zneT\ o = * eT\ + T eT\ + ` eT\ + a z = w, 6

(17)

The values of C1-C6 parameters of Eq. (17) are reported in Table 5 together with their accuracy R2 .

Table 5. Parameters used for the correlation adopted in Eq. (17) and their accuracy R2. ϑ

C1

C2

C3

C4

R2

A

1.333E-02

-5.143E-03

9.095E-04

9.600E-05

0.998

B

105.360

-44.015

-0.888

3.032

1

The reliability of the proposed correlation is shown in Fig. 16.

0.0025

CH4, EQ. [(kg/s)/(kmol/h)]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 42 of 100

25 % C2+ 20% C2+ 15% C2+ 10% C2+ 5% C2+

0.002 0.0015 0.001 0.0005 0 0

0.2

0.4

0.6

0.8

1

z CO2

Figure 16. Reliability of the correlation (Eq. (16)) to represent the specific net equivalent methane as function of the molar fraction of carbon dioxide and ethane in the feed stream when

ACS Paragon Plus Environment

42

Page 43 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

ethane is removed upstream of the dual pressure low-temperature distillation process by extractive distillation.

Though some discrepancies between the results from the proposed correlation and from simulation can be noticed for z CO2 > 0.5 and low C2+ contents, the proposed correlation has been considered valid and it has been used for the overall energy analysis.

3.8 Two-column extractive distillation for the removal of NGLs after Dual Pressure LowTemperature Distillation for methane purification In case of C2+ fraction removal downstream of the dual pressure low-temperature distillation process, a two-column layout has been considered to perform an extractive distillation and break the CO2-C2H6 azeotrope. This solution has been widely adopted in the open literature (see the Introduction section) and has been considered also in this work as state-of-the-art technology. The PFD of the process is shown in Fig. 17.

ACS Paragon Plus Environment

43

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 44 of 100

Figure 17. PFD of the two-columns extractive distillation unit for the separation of the C2+ fraction downstream of the Dual Pressure Low-Temperature Distillation process. → cooling duty required by low-temperature condensers; → heat flow required by reboilers; → amount of useful heat at high temperature that can be recovered inside the process (net equivalent methane savings). For the specific net equivalent methane analysis, cooling duties have been calculated according to Eq. (6) and heat flows have been calculated according to Eq. (1).

The process has been studied with Aspen Hysys V7.3® and the SRK EoS43. The feed stream has been considered at 50 bar and at its bubble point, as the bottom stream of the dual pressure low-temperature distillation process. The stream is a binary mixture of CO2 and C2H6. The molar fraction of CO2 has been varied from 20 to 95 mol%. Under 20 mol% convergence problems

ACS Paragon Plus Environment

44

Page 45 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

occurred. The two distillation columns have 30 theoretical trays and the extraction solvent (nbutane) is fed on the fourth tray from the top of the first distillation column. The pressure of the two distillation units has been set at 35 bar, to remain below the critical pressure of the extraction solvent as considered for the previous case (see Section 3.7). The flow rate of n-butane has been chosen as the minimum value that allows to break the CO2-C2H6 azeotrope assuring high purity of the produced streams, minimizing the leakages of the extraction solvent. CO2 and C2H6 are produced in liquid phase at the top respectively of the first distillation unit and of the extraction solvent regeneration column. The goal is to keep carbon dioxide in liquid phase, in order to use a pump for its further compression. LP steam is used to provide heat at temperatures above 100 °C at the reboilers of the distillation columns. Sensible heat of the hot regenerated n-butane can be recovered inside the process. The results of process simulations for the specific net equivalent methane analysis have been correlated by means of the following expression, function of the molar fraction of CO2 in the feed stream: TA a,A = :xy {0.0135 +|

*

!}

T

/ + 0.2308 +|

*

!}

/ − 7.5564~

(18)

The reliability of the proposed correlation is shown in Fig. 18.

ACS Paragon Plus Environment

45

Industrial & Engineering Chemistry Research

0.002

CH4, EQ. [(kg/s)/(kmol/h)]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 46 of 100

0.0015 0.001 0.0005 0 0

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 z CO2

1

Figure 18. Reliability of the correlation (Eq. (18)) to represent the specific net equivalent methane as function of the molar fraction of carbon dioxide and ethane in the feed stream when ethane is removed downstream of the dual pressure low-temperature distillation process by means of a two-columns extractive distillation unit.

The agreement between simulation results and the correlation has been considered satisfactory for the purposes of the analysis object of this work.

3.9 Three-column extractive distillation for the removal of NGLs after Dual Pressure LowTemperature Distillation for methane purification Alternatively, in order to split the CO2-C2H6 minimum azeotrope, a three-column process has been considered. This process preliminarily removes CO2 or C2H6 (depending on the feed composition) from the azeotrope, reducing the overall energy requirements for the considered separation. According to the isobaric phase diagram of this binary system (Fig. 19) at 35 bar, the

ACS Paragon Plus Environment

46

Page 47 of 100

minimum azeotrope occurs for a CO2 molar fraction of about 0.7. CO2 is the light compound of the mixture.

T [K]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

292 290 288 286 284 282 280 278 276 274 272 270 268 266 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 xCO2

Figure 19. Isobaric phase diagram of the CO2-C2H6 mixture at 35 bar. Calculations have been made using the SRK EoS43.

When its composition (in terms of CO2) is below 0.7, the first distillation column can be used to separate ethane at the bottom and the azeotrope at the top, while for compositions higher than 0.7, the first distillation unit is used to separate CO2 in liquid phase as bottom product stream and the azeotrope at the top. Then, the azeotrope can be separated by means of an extractive distillation unit with n-butane (as for the previous scheme, see Section 3.8). In this way, the energy requirements to separate the azeotrope are reduced since the overall feed stream that enters the extractive distillation unit is lowered by the first distillation unit, which performs a bulk separation. The PFD of the considered solution is shown in Fig. 20.

ACS Paragon Plus Environment

47

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 48 of 100

Figure 20. PFD of the three-columns extractive distillation unit for the separation of the C2+ fraction downstream of the Dual Pressure Low-Temperature Distillation process. → cooling duty required by low-temperature condensers; → heat flow required by reboilers; → amount of useful heat at high temperature that can be recovered inside the process (net equivalent methane savings). For the specific net equivalent methane analysis, cooling duties have been calculated according to Eq. (6) and heat flows have been calculated according to Eq. (1).

The process has been studied with Aspen Hysys V7.3® and the SRK EoS43. The feed stream at its bubble point at 50 bar (as the bottom product from the dual pressure low-temperature distillation process) is a binary CO2-C2H6 mixture. The CO2 content has been varied from 5 up to 95 mol%. The three distillation columns have 30 theoretical trays and are operated at 35 bar. From the top of the first distillation column, the azeotrope is recovered in gas phase and sent to the extractive distillation unit, while the bottom stream contains ethane or carbon dioxide

ACS Paragon Plus Environment

48

Page 49 of 100

(depending on the feed composition) in liquid phase under pressure. The extraction solvent is nbutane and is fed on the fourth tray from the top of the second distillation column. Considerations made for the extractive distillation unit of the two-column scheme (see Section 3.8) are still valid. For energy calculations, since the reboiler of the first distillation column requires heat at low temperature (ambient or lower), its contribution has been neglected, while the two reboilers of the extractive distillation section require heat at temperatures higher than 100 °C, hence LP steam is used. The sensible heat of the hot regenerated extraction solvent is recycled to the process. The results of process simulations, in terms of specific net equivalent methane, have been correlated as function of the inlet CO2 molar fraction of the feed stream, according to: 1.116 × 10.a :xyn4.2603eT o eT < 0.7 `A a,A = (19) T −2.1 × 10.a eT − 2.43 × 10.` eT + 2.68 × 10.` eT > 0.7 The accuracy of the presented correlation is shown in Fig. 21.

0.0025 CH4, EQ: [(kg/s)/(kmol/h)]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

0.002 0.0015 0.001 0.0005 0 0

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 z CO2

1

Figure 21. Reliability of the correlation (Eq. (19)) to represent the specific net equivalent methane as function of the molar fraction of carbon dioxide and ethane in the feed stream when

ACS Paragon Plus Environment

49

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 50 of 100

ethane is removed downstream of the dual pressure low-temperature distillation process by means of a three-columns extractive distillation unit.

The agreement between simulations results and the correlation is good.

3.10 CO2 pumping after low-temperature processes When CO2 recompression for EOR or CCS is considered for the low-temperature purification scheme (Fig. 5), a pump (Fig. 22) can be used since carbon dioxide for this process solution is obtained in liquid phase under pressure.

Figure 22. PFD of the CO2 pump for carbon dioxide recompression in the low-temperature purification chain (Fig. 5). → mechanical power required by the pump, accounted as key parameter for energy analysis (Eq. (7)).

The specific power required by the pump has been calculated using Aspen Hysys V7.3® and the SRK EoS43. Carbon dioxide is available at 50 bar or 35 bar, according to the solution adopted for the C2+ split in the process shown in Fig. (5). When C2+ are removed upstream of the dual pressure low-temperature distillation process, CO2 is produced in liquid phase at 50 bar and the specific pumping power required for its compression to 130 bar is 0.1561 kW per kmol h-1 of CO2. When C2+ are removed downstream of the dual pressure low-temperature distillation

ACS Paragon Plus Environment

50

Page 51 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

process, CO2 is produced in liquid phase at 35 bar and the specific pumping power required for its compression to 130 bar is 0.167 kW per kmol h-1 of CO2.

4. Results and discussion The relative profitability of different process solutions has been discussed on the basis of the methane consumption for energy production. The net equivalent methane of each process has been converted in terms of percentage of produced gas that has to be burned to provide the energy for the process: in this way, the profitability is established on the basis of how much product can be effectively sold to the market, allowing to study also the feasibility of each standalone process according to the CO2 and C2+ content in the feed gas.

4.1 CO2-C2H6 azeotrope separation by means of two vs three distillation columns As first step, two solutions for the separation of the CO2-C2H6 minimum azeotrope have been compared: a classic two-column extractive distillation and a three-column process, which includes a bulk removal of the azeotrope and, subsequently, an extractive distillation. The results of the comparison are shown in Fig. 23.

ACS Paragon Plus Environment

51

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 52 of 100

Figure 23. Energy comparison between (─) a two-column extractive distillation unit and (─) a three-column extractive distillation process for the split of the CO2-C2H6 minimum azeotrope (─: isobaric phase diagram of the CO2-C2H6 mixture at 35 bar, calculated with the SRK EoS43.

There is a trade-off between the two proposed process solutions: for values of the inlet CO2 molar fraction below 0.48 and above 0.75, the three-column process results to be less energy intensive than the classical two-column process, while for values around the azeotrope composition (about 0.7), the classic two-column process is less energy intensive. It is possible to notice, moreover, that, for values of the CO2 molar fraction greater than 0.5, the energy consumptions of the two-column extractive distillation process is not very sensitive to the inlet content of carbon dioxide. The feasibility area of Fig. 23 has been considered to account for the specific net equivalent methane of the CO2-C2H6 separation unit downstream of the dual pressure low-temperature distillation process. Eq. (18) is considered for z CO2 between 0.48 and 0.75, while Eq. (19) is used below 0.48 and above 0.75.

ACS Paragon Plus Environment

52

Page 53 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

4.2 Downstream vs upstream C2+ removal in the low-temperature purification plant A second step of the present work has been the energy comparison between the recovery of the C2+ fraction of natural gas upstream or downstream of the low-temperature purification process (Fig. 5). In literature, as discussed in the Introduction Section, different works have been proposed on process solutions for the integration between low-temperature natural gas purification processes and NGLs recovery, but no work has been found discussing the relative profitability of different schemes on the basis of a model-based approach and process simulations results. In this work, a numeric comparison is shown. For this analysis, only the energy costs of the dual pressure low-temperature distillation process and the ones of the C2+ recovery units (upstream or downstream) have been considered, since desulfurization and dehydration steps are the same for both the cases in the low-temperature chain. For the downstream recovery of NGLs both the two-columns and the three-columns scheme have been considered, choosing the best solution as for energy performances, according to the relative profitability for different CO2 contents. For this study, the inlet composition of CO2 has been varied from 5 to 65 mol% and the C2H6 one from 5 to 25 mol%. Results are shown in Fig. 24.

a)

ACS Paragon Plus Environment

53

Industrial & Engineering Chemistry Research

80 Produced CH4 Burned %

70

5% C2H6

60 50 40 30 20 10 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 z CO2

b) 90 80

Produced CH4 Burned %

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 54 of 100

10% C2H6

70 60 50 40 30 20 10 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 z CO2

c)

ACS Paragon Plus Environment

54

100 90 80 70 60 50 40 30 20 10 0

15% C2H6

0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 z CO2

d)

Produced CH4 Burned %

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Produced CH4 Burned %

Page 55 of 100

130 120 110 100 90 80 70 60 50 40 30 20 10 0

20% C2H6

0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 z CO2

e)

ACS Paragon Plus Environment

55

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Produced CH4 Burned %

Industrial & Engineering Chemistry Research

220 200 180 160 140 120 100 80 60 40 20 0

Page 56 of 100

25% C2H6

0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 z CO2

Figure 24. Energy comparison, in terms of % of produced methane that has to be burned to provide energy to the process, between (─) Dual Pressure Low-Temperature Distillation process with downstream C2+ recovery and (─) Dual Pressure Low-Temperature Distillation process with upstream C2+ recovery. a) 5 mol% of C2H6, b) 10 mol% of C2H6, c) 15 mol% of C2H6, d) 20 mol% of C2H6 and e) 25 mol% of C2H6 in the feed stream.

For an inlet feed stream having 5 mol% of ethane, the downstream removal of the C2+ fraction is always less energy intensive than the upstream one. The difference is enhanced above 20 mol% of CO2 in the inlet feed. When the ethane content is increased to 10 mol%, a trade-off between the two process solutions occurs for an inlet CO2 content between 30 and 35 mol%. The downstream separation of C2+ is favored, in this case, when the inlet content of CO2 is higher than 35 mol%. For a feed stream containing 15 mol% of ethane, the trade-off is shifted up when CO2 in the feed is about 50 mol%. For feed streams having 20 and 25 mol% of ethane, the upstream removal of the C2+ fraction is always more profitable for any content of carbon dioxide in the feed gas.

ACS Paragon Plus Environment

56

Page 57 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Considering the downstream CO2-C2H6 separation scheme, for 5 mol% of inlet ethane, the two-column extractive distillation process is less energy intensive for CO2 contents from 5 mol% to 15 mol%, otherwise the three-column extractive distillation process is favored. At 10 mol% of ethane in the feed gas, the two-column process is more profitable for CO2 contents of the feed stream between 10 and 30 mol%. Increasing the ethane content from 15 to 25 mol%, the twocolumn solution for the separation of CO2 and C2H6 downstream of the low-temperature purification process is favored for CO2 contents of the feed stream between 15 and 45, 20 and 60, 25 and 65 mol% respectively. Moreover, the purification of natural gas streams with high CO2 and C2+ contents results to be extremely energy intensive, with overall requirements higher than 100% of the produced methane.

4.3 Classic vs low-temperature purification plants without CO2 recompression The energy analysis has been extended to the complete natural gas purification plant, neglecting the recompression of CO2 in a preliminary phase. A classic scheme (Fig. 4) has been compared to a low-temperature scheme (Fig. 5) for the purification of natural gas and the recovery of the C2+ fraction. Energy expenses have been accounted, after the net equivalent methane analysis, as percent of the produced gas that has to be burned to provide energy to the overall purification process. For the low-temperature purification scheme, both upstream and downstream C2+ removal has been considered, according to the results obtained previously (see Section 4.2). The results of the comparison between the two complete purification plants, without CO2 recompression, are shown in Figs. 25-28. As for the feed gas composition, the H2S content has been varied from 0 to 15 mol%, the C2H6 content from 0 to 25 mol% and the CO2 content from 5 to 65 mol%.

ACS Paragon Plus Environment

57

Industrial & Engineering Chemistry Research

a) 60 0% C2+ % CH4 Burned

50 40 30 20 10 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

b) 70 5% C2+

60

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 58 of 100

50 40 30 20 10 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

c)

ACS Paragon Plus Environment

58

90 80 70 60 50 40 30 20 10 0

10% C2+

5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

d) 120 15% C2+ 100

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

% CH4 Burned

Page 59 of 100

80 60 40 20 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

e)

ACS Paragon Plus Environment

59

Industrial & Engineering Chemistry Research

160 20% C2+

% CH4 Burned

140 120 100 80 60 40 20 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

f) 250 25% C2+ 200

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 60 of 100

150 100 50 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

Figure 25. Energy comparison, in terms of % of produced methane that has to be burned to provide energy to the process, between (─) classic natural gas purification chain (Fig. 4) and (─) low-temperature purification chain (Fig. 5) for a gas feed stream having 0 mol% of H2S and a) 0 mol% of C2H6, b) 5 mol% of C2H6, c) 10 mol% of C2H6, d) 15 mol% of C2H6, e) 20 mol% of C2H6 and f) 25 mol% of C2H6.

ACS Paragon Plus Environment

60

Page 61 of 100

When no H2S and no C2+ are present in the feed stream, results are in agreement with the ones obtained in a previous work14: the breakeven point between the two technologies occurs for a CO2 content of about 8 mol%. When the C2+ content in the raw gas is increased from 5 to 25 mol%, the breakeven point occurs for a CO2 content of the raw gas between 20 and 35 mol%. It is possible to notice that, the increasing content of the C2+ fraction in the raw gas feed tends to reduce the difference between the energy expenses of the two processes and the overall energy expenses may become greater than 100% of the produced methane. The low-temperature purification plant is always favored than the classical one for high concentrations of CO2 in the inlet gas.

a) 80 0% C2+

70

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

60 50 40 30 20 10 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

b)

ACS Paragon Plus Environment

61

Industrial & Engineering Chemistry Research

100 5% C2+

% CH4 Burned

80 60 40 20 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

c) 120 10% C2+ 100

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 62 of 100

80 60 40 20 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

d)

ACS Paragon Plus Environment

62

Page 63 of 100

160 15% C2+

% CH4 Burned

140 120 100 80 60 40 20 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

e) 250 20% C2+ 200

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

150 100 50 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

f)

ACS Paragon Plus Environment

63

Industrial & Engineering Chemistry Research

500 25% C2+ 400

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 64 of 100

300 200 100 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

Figure 26. Energy comparison, in terms of % of produced methane that has to be burned to provide energy to the process, between (─) classic natural gas purification chain (Fig. 4) and (─) low-temperature purification chain (Fig. 5) for a gas feed stream having 5 mol% of H2S and a) 0 mol% of C2H6, b) 5 mol% of C2H6, c) 10 mol% of C2H6, d) 15 mol% of C2H6, e) 20 mol% of C2H6 and f) 25 mol% of C2H6.

The same considerations hold for an inlet gas having 5 mol% of H2S. The presence of hydrogen sulfide increases the overall energy expenses, but the obtained results in terms of breakeven points (BEPs) and trends are similar.

a)

ACS Paragon Plus Environment

64

Page 65 of 100

100 0% C2+

% CH4 Burned

80 60 40 20 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

b) 140 5% C2+

120

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

100 80 60 40 20 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

c)

ACS Paragon Plus Environment

65

180 160 140 120 100 80 60 40 20 0

Page 66 of 100

10% C2+

5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

d) 250 15% C2+ 200

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

% CH4 Burned

Industrial & Engineering Chemistry Research

150 100 50 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

e)

ACS Paragon Plus Environment

66

Page 67 of 100

500 20% C2+

% CH4 Burned

400 300 200 100 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

f) 500 25% C2+ 400

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

300 200 100 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

Figure 27. Energy comparison, in terms of % of produced methane that has to be burned to provide energy to the process, between (─) classic natural gas purification chain (Fig. 4) and (─) low-temperature purification chain (Fig. 5) for a gas feed stream having 10 mol% of H2S and a) 0 mol% of C2H6, b) 5 mol% of C2H6, c) 10 mol% of C2H6, d) 15 mol% of C2H6, e) 20 mol% of C2H6 and f) 25 mol% of C2H6.

ACS Paragon Plus Environment

67

Industrial & Engineering Chemistry Research

For the case with 10 mol% of H2S the BEPs are shifted to higher CO2 contents: for instance for 25 mol% of C2+ in the raw gas feed the BEP is shifted from 35 mol% to 40 mol% of CO2 in the feed stream.

a) 140 0% C2+

% CH4 Burned

120 100 80 60 40 20 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

b)

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 68 of 100

180 160 140 120 100 80 60 40 20 0

5% C2+

5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

c)

ACS Paragon Plus Environment

68

Page 69 of 100

300 10% C2+

% CH4 Burned

250 200 150 100 50 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

d) 600 15% C2+ 500

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

400 300 200 100 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

e)

ACS Paragon Plus Environment

69

Industrial & Engineering Chemistry Research

500 20% C2+

% CH4 Burned

400 300 200 100 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

f) 500 25% C2+ 400

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 70 of 100

300 200 100 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

Figure 28. Energy comparison, in terms of % of produced methane that has to be burned to provide energy to the process, between (─) classic natural gas purification chain (Fig. 4) and (─) low-temperature purification chain (Fig. 5) for a gas feed stream having 15 mol% of H2S and a) 0 mol% of C2H6, b) 5 mol% of C2H6, c) 10 mol% of C2H6, d) 15 mol% of C2H6, e) 20 mol% of C2H6 and f) 25 mol% of C2H6.

ACS Paragon Plus Environment

70

Page 71 of 100

For a gas feed with 15 mol% of H2S, it is possible to notice that, for gases having also a high amount of C2+, the classic process scheme is always less energy intensive than the lowtemperature one up to 40 mol% of CO2 in the feed gas. The presence of H2S and C2+ increases the overall energy expenses and reduces the differences between energy consumptions of the considered schemes, but it can be noticed that for C2+ content ≤ 10 mol% the low-temperature process allows to significantly reduce the energy expenses below 100% of burned product even when the classical scheme requires more than 100% of the produced gas.

4.4 Classic vs low-temperature purification plants integrated with EOR or CCS The two process solutions have been finally compared when CO2 recompression for EOR or CCS is required (Figs. 4-5). The results are shown here in Figs. 29-32.

a) 70 0% C2+

60

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

50 40 30 20 10 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

b)

ACS Paragon Plus Environment

71

Industrial & Engineering Chemistry Research

80 5% C2+

% CH4 Burned

70 60 50 40 30 20 10 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

c) 100 10% C2+ 80

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 72 of 100

60 40 20 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

d)

ACS Paragon Plus Environment

72

Page 73 of 100

120 15% C2+

% CH4 Burned

100 80 60 40 20 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

e) 160 20% C2+

140

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

120 100 80 60 40 20 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

f)

ACS Paragon Plus Environment

73

Industrial & Engineering Chemistry Research

250 25% C2+

% CH4 Burned

200 150 100 50 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

Figure 29. Energy comparison, in terms of % of produced methane that has to be burned to provide energy to the process, between (─) classic natural gas purification chain (Fig. 4) and (─) low-temperature purification chain (Fig. 5), with CO2 recompression, for a gas feed stream having 0 mol% of H2S and a) 0 mol% of C2H6, b) 5 mol% of C2H6, c) 10 mol% of C2H6, d) 15 mol% of C2H6, e) 20 mol% of C2H6 and f) 25 mol% of C2H6.

a)

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 74 of 100

90 80 70 60 50 40 30 20 10 0

0% C2+

5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

ACS Paragon Plus Environment

74

Page 75 of 100

b) 100 5% C2+

% CH4 Burned

80 60 40 20 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

c) 140 10% C2+

120

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

100 80 60 40 20 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

d)

ACS Paragon Plus Environment

75

180 160 140 120 100 80 60 40 20 0

Page 76 of 100

15% C2+

5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

e) 300 20% C2+ 250

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

% CH4 Burned

Industrial & Engineering Chemistry Research

200 150 100 50 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

f)

ACS Paragon Plus Environment

76

Page 77 of 100

600 25% C2+

% CH4 Burned

500 400 300 200 100 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

Figure 30. Energy comparison, in terms of % of produced methane that has to be burned to provide energy to the process, between (─) classic natural gas purification chain (Fig. 4) and (─) low-temperature purification chain (Fig. 5), with CO2 recompression, for a gas feed stream having 5 mol% of H2S and a) 0 mol% of C2H6, b) 5 mol% of C2H6, c) 10 mol% of C2H6, d) 15 mol% of C2H6, e) 20 mol% of C2H6 and f) 25 mol% of C2H6.

a) 120 0% C2+ 100

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

80 60 40 20 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

ACS Paragon Plus Environment

77

Industrial & Engineering Chemistry Research

b) 140 5% C2+

% CH4 Burned

120 100 80 60 40 20 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

c)

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 78 of 100

180 160 140 120 100 80 60 40 20 0

10% C2+

5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

d)

ACS Paragon Plus Environment

78

Page 79 of 100

300 15% C2+

% CH4 Burned

250 200 150 100 50 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

e) 600 20% C2+ 500

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

400 300 200 100 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

f)

ACS Paragon Plus Environment

79

Industrial & Engineering Chemistry Research

500 25% C2+

% CH4 Burned

400 300 200 100 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

Figure 31. Energy comparison, in terms of % of produced methane that has to be burned to provide energy to the process, between (─) classic natural gas purification chain (Fig. 4) and (─) low-temperature purification chain (Fig. 5), with CO2 recompression, for a gas feed stream having 10 mol% of H2S and a) 0 mol% of C2H6, b) 5 mol% of C2H6, c) 10 mol% of C2H6, d) 15 mol% of C2H6, e) 20 mol% of C2H6 and f) 25 mol% of C2H6.

a) 140 0% C2+

120

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 80 of 100

100 80 60 40 20 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

ACS Paragon Plus Environment

80

Page 81 of 100

b) 200

% CH4 Burned

5% C2+ 150 100 50 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

c) 300 10% C2+ 250

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

200 150 100 50 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

d)

ACS Paragon Plus Environment

81

Industrial & Engineering Chemistry Research

600 15% C2+

% CH4 Burned

500 400 300 200 100 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

e) 600 20% C2+ 500

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 82 of 100

400 300 200 100 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

f)

ACS Paragon Plus Environment

82

Page 83 of 100

500 25% C2+ 400

% CH4 Burned

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

300 200 100 0 5

10 15 20 25 30 35 40 45 50 55 60 65 zCO2 [mol%]

Figure 32. Energy comparison, in terms of % of produced methane that has to be burned to provide energy to the process, between (─) classic natural gas purification chain (Fig. 4) and (─) low-temperature purification chain (Fig. 5), with CO2 recompression, for a gas feed stream having 15 mol% of H2S and a) 0 mol% of C2H6, b) 5 mol% of C2H6, c) 10 mol% of C2H6, d) 15 mol% of C2H6, e) 20 mol% of C2H6 and f) 25 mol% of C2H6.

The trend of the obtained results for the two considered natural gas production schemes with CO2 recompression are qualitatively similar to the ones obtained previously. However, the CO2 recompression affects the breakeven points, shifting them at lower values of the inlet CO2 molar fraction. The highest penalty in terms of overall energy consumptions is paid for the recompression and dehydration of the wet CO2 stream coming from the regeneration column of the acid gas removal unit, which operates a chemical absorption of acidic gases in aqueous MDEA. The CO2 is rejected in gas phase at atmospheric pressure from traditional MDEA unit, while, from low-temperature natural gas distillation processes, it is rejected typically in liquid phase under pressure. Hence, the recompression work required by the classic purification plant is

ACS Paragon Plus Environment

83

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 84 of 100

higher than the one required by the low-temperature plant. In this way, when CO2 recompression is considered, the overall energy consumption of the classic scheme is significantly increased respect to the one of the low-temperature scheme. In this way, the breakeven points for any amount of H2S and C2+ in the feed stream are moved to lower contents of CO2 in the raw gas and the low-temperature purification scheme is favored. In order to give a more general overview of the obtained results and the cost of CO2 recompression on energy expenses for the two natural gas purification schemes, the main results of the study have been summarized in Table 6.

Table 6. Overall results of the energy analysis and comparison between the classic natural gas purification chain and the low-temperature purification chain with and without CO2 recompression. CO2-C2H6 separation process or when the LT process is more profitable

CO2 mol% in the feed at which 100% of the produced gas has to be burned to produce energy for the process

Feed

BEP without CO2 BEP with CO2 C2+ removal upstream (U) recompression recompression downstream (D) in LT scheme

H2S C2+

CO2 mol%

CO2 mol%

U/D

2/3 columns

Classic LT Scheme Scheme

0

8

8

N/A (Not Applicable)

N/A

N/A

N/A

5

20

18

D (Figure 5b,d)

3

N/A

N/A

10

35

30

U (Figure 5a,c) at BEP, then D (Figure 3 5b,d)

N/A

N/A

15

35

25

U (Figure 5a,c) from BEP to 55 mol% 3 of CO2, then D (Figure 5b,d)

65

N/A

20

30

25

U (Figure 5a,c)

N/A

60

65

25

35

30

U (Figure 5a,c)

N/A

55

58

0

8

5

N/A

N/A

N/A

N/A

0

5

ACS Paragon Plus Environment

84

Page 85 of 100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

5

20

18

D (Figure 5b,d)

10

35

25

15

30

20

3

N/A

N/A

U (Figure 5a,c) from BEP to 35 mol% 3 of CO2, then D (Figure 5b,d)

63

N/A

23

U (Figure 5a,c) from BEP to 50 mol% 3 of CO2, then D (Figure 5b,d)

58

63

28

23

U (Figure 5a,c)

N/A

53

58

25

38

28

U (Figure 5a,c)

N/A

51

53

0

8